List your company

Tool Retrieval on TOOLRET Zero-Shot Macro-Avg

36.9nDCG@10TOOLQP
19.2223.8128.432.99Jan 12, 2026

Evaluation results

MethodLinks
TOOLQPMethod Category=Fine-tuning, Training Data Size=10k, Backbone Model=1.7B2026.0136.935.8
bge-gemmaMethod Category=Re-ranking with cross-encoder2026.013429.7
Q2P/SFTMethod Category=Fine-tuning, Training Data Size=10k, Backbone Model=1.5B/1.7B models2026.0133.231.1
Q2E/ZSMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.013229.8
gte-Qwen/ContrastiveFTMethod Category=Fine-tuning, Training Data Size=10k, Backbone Model=1.5B/1.7B models2026.0131.529
Q2D/PRFMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.0131.227.9
Q2E/PRFMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.013128.4
Re-InvokeMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.0130.327.9
bge-m3Method Category=Re-ranking with cross-encoder2026.0130.126.8
Base Retriever (gte-Qwen)Method Category=Base Retriever2026.0129.827
D2QMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.0128.426.4
HyDE/PRFMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.0127.624.4
Q2D/ZSMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.0126.726.9
TOOLQP-FORMATMethod Category=Fine-tuning, Training Data Size=10k, Backbone Model=1.7B2026.0126.125.9
HyDE/ZSMethod Category=Prompting, Backbone Model=Qwen3-30B-A3B-Instruct2026.0119.921.8